AN ACOUSTIC ANOMALY

THE "CORRECTED" ACOUSTIC OF SAN FRANCISCO'S DAVIES HALL

 

©1983 The Anstendig Institute

Revised 1984

 

This is one of a series of papers by The Anstendig Institute on acoustics and sound equalization. We recommend that these papers be read together. The paper "Concert Hall Acoustics" is basic in that it explains the usually misunderstood fundamentals of acoustics (volume, equalization, and reverberation).

 

Modern acoustical engineering is a young field using new techniques. Opportunities for testing those techniques in the design of a concert hall are necessarily limited. Every new hall is an experiment and it must be expected that results might not immediately meet desired objectives, especially when, as in Davies Hall, a new design was demanded that would allow the hall to serve many purposes.

The Anstendig Institute has faced the same dilemma that all responsible people face when they have to choose the right attitude toward a situation as sensitive as the controversy surrounding the acoustic of Davies Hall. Careful consideration has led us to believe that the only valid attitude is to be truthful and insist that the hall is faulty, that music-making in the present hall will never reflect the true capabilities of the performing artists, and that there is a distinct danger in subjecting the broad musical public to these performances.

 

The public is becoming tuned to an undesirable level of art and will eventually lose its powers of discrimination if it has not done so already. Reports in the newspapers of huge amounts of applause for second-rate performances indicate that the public already is conditioned to respond to music on a culturally low level. The only way to be uplifted into higher levels of discrimination and artistic experience, which is the purpose of art, is to be exposed to performances of finer quality. As sad as it makes us to say it, this amounts to a condemnation of the present Davies Hall because in that hall performers capable of such performances cannot achieve their full potential.

I. THE SOUND

The corrections in San Francisco's Davies Hall have resulted in a curious acoustical anomaly: the use of new techniques and materials in the original hall and the subsequent corrective procedures has resulted in a sound unlike anything that has existed before.

 

Originally, Davies Hall had three main problems: l) exaggeratedly loud high frequencies in the most sensitive range of our hearing (from ca 2000 Hz to 4000 Hz); 2) a thickness in the sound in the musical middle register (ca 300 Hz to lO0O Hz) due to an over-prominence of the first harmonic overtones1; and, 3) an over-abundance of reflected sound, the quality of which is usually mistakenly called reverberation2 These are essentially the same problems that occur in unequalized sound-reproduction and, in fact, the sound in Davies Hall originally resembled typically imperfect, distorted, modern sound reproduction.3 A fourth problem, an overly live auditorium, was quickly improved by the installation of a carpet, but the auditorium still remains very live.

 

The high frequency problem has been reduced either by the latest corrections of the hall and/or by adjustments the orchestra has made in their manner of playing. But the second and third problems are now more prominent, giving the sound a curiously distorted, almost grotesque character which still resembles the sound of typical, flawed, unequalized sound­ reproduction, but with the high-frequency tone controls turned down.4

 

The apparent reduction in the high frequencies has reduced the volume of the grating peaks in those high frequencies which we hear loudest (ca 2000 Hz to 5000 Hz). These peaks cause the most distraction and tension, keeping the listener from relaxing enough to hear the expressive content of a performance. But subduing the higher frequencies has given added prominence to the frequencies from ca l000 Hz to 1900 Hz. This gives the sound an apparent euphoric quality which is in reality a distortion that still degrades the expressive content. This illusory euphoric quality becomes tiring for two interrelated reasons: l) it is really an exaggeration of the first harmonics, particularly those of the strings, a distortion that falsifies the sound of the instruments and muddies the sound textures; 2) these overly loud harmonics make most of the music, particularly the sound of the string instruments, monotonously unvaried in quality and expression.

 

It is the expression in music that best sustains our interest. A player varies expression by changing the quality and dynamics of the instrumental tone. Since overtones are excited by fundamental tones and not vice versa, the expressive changes occur principally in the fundamental notes (the actual written notes), not in their harmonics. As mechanical-physical phenomena caused by the fundamental tone, some of the overtones necessarily occur ever-so-slightly after the fundamental has been produced. The performer can change the dynamic balance of the overtones relative to each other by changing the manner in which the tone is produced, but the expressive content--that which makes one cry, laugh, become pensive, or otherwise react expressively to the music--is conveyed by the fundamental tones.

 

In Davies Hall the overtones are too loud in relation to the fundamentals. As a result, the overtone structure has become the dominant factor coloring the sound, obliterating the expressive tonal subtleties carried by the fundamentals. This imparts a curiously expressionless quality to the music-making. While the euphoric quality due to the overtones between l000 and 2000 Hz can at first seem lovely to the inexperienced ear, the sameness and lack of differentiation in tone and expression causes loss of interest and fatigue, but in a subtler, more subliminal way than the original grating high-frequency peaks. The extreme high frequency edginess was an immediately apparent irritation but the present problems take time to discern. One could at first have the impression that the hall has been vastly improved, even fully corrected, and many people have that impression. But they are wrong.

 

The apparent reduction in the high-frequency edginess, which was most noticeable in the strings, may result more from the orchestra members adjusting their tone to the peculiarities of the hall than from any fine-tuning of the hall itself. Originally, only the high frequencies between 2000 Hz and 5000 Hz were too loud, forming a peak that stuck out in relation to the other frequencies. But the apparent changes in the sound of the hall have reduced the high frequencies without eliminating the peaks. If the changes had been in the hall, the high frequencies would have been reduced in all instruments. Instruments like the violins, whose tone can be changed over a wide range, have improved more than the cymbals, triangle, celesta, and other instruments whose sounds range in frequencies well above 4000 Hz and cannot as readily modify their frequency content. The manner of attack and articulation of many instruments have also changed. Therefore, many of the apparent changes in the sound come from the players, not the hall, which limits the freedom of the orchestra to play in whatever manner the music demands.

 

Concerts by visiting groups who have their own set way of playing and are not familiar enough with the hall to adjust to it, seem to bear out this point. A concert by the Amadeus Quartet, a relatively mellow sounding group with its own very strongly established manner of playing, exhibited the original problems of the hall, including a harsh edginess.

 

The original edginess in tone, especially in the violins, still occurs when the instruments 'are played loudly. This shows that, whether accomplished by the players or by the acoustician, the reduction in the high frequencies has not corrected the peaks. The reason the peaks suddenly become apparent at particular volume levels is due to a threshold effect in the way we hear the higher frequencies, i.e., we do not hear them unless they are louder than a specific volume level.5 Below these threshold levels, they simply are not registered.

 

The second and third problems are interrelated. The over-abundance of harmonics is the result of changes in the equalization due to overly loud and overly abundant reflected sound. The problem of the overly loud first harmonics is worse in those instruments at the back of the stage where a curious sound quality due to the over-abundance of reflected sound is most pronounced. The sound is muddied by the reflections, but especially by the resulting exaggerated first harmonics. The sounds of the horns, tympani, and tuba are exaggeratedly loud and diffuse due to the strong reflections. They also suffer particularly from the thick, muddied quality. The exaggerated first order harmonics very nearly obliterate the fundamental tones. In many ways, the tympani and tuba sound like they are being reproduced by loudspeakers that are too small to reproduce much bass below l00 Hz. with such speakers, one hears mainly the first harmonics of the bass instruments and not the bass tones themselves. Many speakers are actually designed to exaggerate this frequency range in order to give the listener the impression of deep bass fundamentals which the speaker is, in fact, not able to produce. The world of sound reproduction is full of such tricks which have unfortunately conditioned most people's ears to accept sound qualities that are gross distortions. The design of Davies Hall intentionally tries to translate these electronic qualities into acoustical form.

 

In Davies Hall, the sound of the instruments is diffuse and unfocused. Since this is most evident with instruments sitting closer to the back wall of the stage, it would follow that much of the problem is caused by that wall. But the instruments that suffer most, the tuba and tympani, radiate upward, indicating that the reflectors hanging from the ceiling are also part of the problem. The diffuse, unfocused nature of the sound is due to the slightly convex shape of the reflectors and to the convex shapes of many of the reflecting surfaces of the hall6. (The reflectors are not precisely named. They are really diffusers, not reflectors. Reflectors would have to be flat or concave.)

 

More "reflectors" have been added to the original design. There are now so many so-called reflectors so close together over the stage that they in effect amount to a dropped ceiling, but one that diffuses rather than conducts the sound as a ceiling directly over an orchestra normally would. Also, what is in effect a new, lower ceiling in many ways negates the reasons for the original shape of the hall. Why should there now be the extremely high ceiling above the reflectors? Since the reflectors are corrective procedures that were included in the design of the hall, the hall itself was purposely designed with basic flaws that the reflectors were supposed to correct. But it was evidently not understood that reflecting surfaces cannot correct sound. Unlike a mirror reflecting light waves, sound-reflecting surfaces change the balance of frequencies (the equalization), reflecting the different frequencies unequally, distorting the sound instead of improving it.

 

The added reflectors may be the real cause for the change in the manner in which the instruments are played. With the additional reflected sound onstage, the orchestra itself hears more of the high-frequency peaks that the audience hears. They therefore both consciously and subconsciously adjust their tone to produce as little harshness as possible. Obviously whatever they do to compensate for the problems of the hall, their playing is limited by doing so.

Because any reflecting surface changes the equalization of sound, the use of a relatively low ceiling above an orchestra even as a means of conducting the sound, should probably be avoided. Opera houses with no ceiling above the stage and often with sets that have no effective reflecting surfaces continuously demonstrate that reflecting surfaces, whether side-walls or ceilings, are not necessary.

From the point of view of simple, basic physics, the highly polished, strongly reflecting wall so close behind the orchestra is an anomaly of design and a major cause of the problems of Davies Hall. Any wall has to reflect the sound of instruments nearer to it much more strongly than that of instruments farther away. In Davies Hall the closer to the wall the instruments are placed, the louder they sound. The instruments directly in front of the reflecting wall will sound louder than the rest of the orchestra. This condition is compounded because the instruments that are the loudest anyway occupy this position. This physical-mathematical relationship of volume in relation to distance is enormously exaggerated on the stage because of the great difference between the distance from the wall to the players directly in front of the wall and the distance from the wall to the players at the front of the stage. Those players far from the wall hear the instruments near the wall proportionally louder than those instruments farther away from it. Thus the violinists do not hear the cellos well but they hear the horns and tympani overly loud, and the horn players hardly hear the violins and cellos in comparison with the instruments around them.

 

The basic difference in volume is not the only problem caused by the rear wall. The players in front of the stage hear a different proportion of reflected and direct sound from their own instruments than is heard by those players close to the reflecting wall. Reflected sound has a different overtone structure than direct sound; thus, the sound heard by players in different locations on the stage does not have the same equalization characteristics. In other words, according to their location, each player hears the other instruments with a different characteristic sound quality. If such a wall was necessary, the players at the back of the orchestra should have been able to sit much farther away from it so that there would be a smaller ratio between their distance from the wall and the depth of the stage.

 

Such a strongly reflective wall behind the orchestra is probably not necessary, especially with part of the audience sitting behind the orchestra. Excellently balanced concerts have been played in halls with a canvas backdrop behind the orchestra, and canvas absorbs sound. Jean Morel, the well-known conducting teacher at Juilliard, who had an extraordinary ear for balances, happily conducted concerts on just such a stage that had only curtains and no reflecting surfaces at all in the wings at the sides of the stage, and no reflecting ceiling above it.

 

It is strangely fascinating to hear the distorted sound of the orchestra in Davies Hall. Now, with less edgy high-­frequency peaks, one can more easily listen, but one soon realizes that the distortions all stand in the way of experiencing the music, and impede the music-making itself. The exaggeratedly overtone-rich sound of the horns, the blubbery, overly resonant sound of the tuba, and the badly defined, diffuse, resonant sound of the tympani are sounds one has never before heard from these instruments. The anomaly of this hall is that these sounds are unnatural. They are manufactured by the maverick design of the hall and occur nowhere else. Due to the reflecting wall, the instruments in back sound larger than life and distorted in relation to the rest of the orchestra, the sounds of the separate sections of the orchestra cannot fuse, and the performers are distracted by a sound that they know is wrong. The player's attention is directed to trying to correct the sound and thus diverted from the expressive content of the performance.

 

I. THE PLAYERS

The problems of Davies Hall have caused the players of the San Francisco Symphony to change their technique of playing in an attempt to compensate for the anomalies that they hear. In order to compensate for the acoustic faults of the hall, they have to play in a technically wrong manner. Interestingly, the ensemble that, in my experience, sounded best in Davies Hall, the Academy of Saint Martin in the Fields, is one of the first to use similar faulty techniques. This group spends much time in the recording studio, evidently evaluating its sound ­quality from its recordings and adjusting its manner of playing to the distortions of the sound reproduction. Also, the Academy orchestra is small enough to sit towards the front of the stage, avoiding the problems near the wall--something that a large orchestra cannot do.

 

The tendency of more and more musicians to unwittingly judge their playing from the playback of their recordings has led to the use of many technically wrong methods of producing sound, the most often encountered of which is a trademark of the Academy's string playing. Sound reproduction suffers particularly from exaggerated high frequency peaks. In order to subdue the high-frequency peaks that they hear in the playback, the Academy strings play their instruments with less "bite", which is accomplished by using "more bow and less pressure" (i.e., longer, faster movements of the bow and less pressure of the bow on the strings). This is a known technique that is called for in certain passages of music, but most of the time it is out of place and just plain wrong. It derives from a misunderstanding of techniques sometimes used by the extraordinary conductor Herbert von Karajan, whose conducting courses have influenced whole generations of musicians.

 

Maestro Karajan for some time experimented with developing more beauty of sound and eliminating ugliness from an orchestra's manner of playing. This is often accomplished by softening the orchestra's initial attack (the manner in which the players start the tone) and by making extensive use of the technique of less pressure and more bow. But what works for a Karajan does not work for others. He possesses an extraordinary dynamic energy which allows him to elicit strong, even enormous, attacks and accents from his orchestra even when played in this cushioned manner. And Karajan will demand a straight, hard sound when it is called for. That adaptations and misuse of this technique, which the San Francisco Symphony has also begun to use, sound well in a faulty hall like Davies Hall does not make the technique or the hall right. In Davies Hall, using this technique amounts to a subtle, sophisticated compounding of errors. The hall should be corrected and not the manner of playing distorted to compensate for the problems of the hall.

 

It has become difficult to tell how much of the improvement in the original high frequency peaks is due to changes in the hall and how much is due to changes in the orchestra's playing. Therefore, any further corrections should be made after evaluating the sound of another orchestra that is not used to the hall, i.e., not already conditioned to compensate for the problems of the hall in their manner of playing.

 

There is much talk about the fact that the musicians cannot "hear themselves", but there is little specific description of what that statement means. Can they not hear their own instruments? Can they not hear their fellow players? Do they hear themselves but in a distorted way? Discussion with members of the orchestra has clarified that they can hear themselves but that they have difficulty hearing certain of the other instrumental groups. Which groups cannot be heard varies with the particular player's location. But most members have mentioned not being able to hear instruments like the violins and cellos that are towards the front of the stage. Obviously this confirms the above mentioned point that the instruments are not equally loud, with the sound from those instruments closer to the back wall louder than that of the instruments farther away from it.

 

But the most important consideration is that, although they can hear themselves, the players, particularly those at the rear of the stage, do not like what they hear. The sound comes back to them sounding wrong, i.e., neither the way they thought they were playing it, nor the way they know their instruments should sound. Thus, an indefensible amount of their energy and attention is devoted to trying to compensate by changing the manner in which they play their instruments. Since most classical music is written so that it utilizes the utmost of an instrument's capabilities, the music has to suffer, and greatly so, when the players are thus distracted. As a permanent situation, this is intolerable.

 

The San Francisco Symphony is expected to sustain first class levels of music-making. The musicians spent their whole lives perfecting their playing in order to participate in the most wondrous experiences they know of, and now find themselves in a situation where there is an insurmountable obstacle that keeps those experiences from happening. One would do well to heed the orchestra's opinion and not believe those who claim that the hall is really acceptable, even quite good. The players know best what they should sound like and when a situation is or is not conducive to the real musical experience. They know that this hall is faulty, making it difficult to impossible for them to playas well as they should...a demoralizing situation that makes it imperative to take drastic measures to provide them with an acoustically sympathetic environment.

 

Much of the problem of Davies Hall seems to be that the acoustician has a wrong concept of correct sound quality. The orchestra members relate that, while corrections were being made, he often joined them on stage, and whenever they pointed out to him what was wrong with the sound, he would tell them they were wrong and the sound was the way it should be. Obviously the acoustician's idea of what an orchestra sounds like is different from that of the players. But the players know how they are playing their instruments and how the instruments sound in all types of other acoustical surroundings. It is obvious that the acoustician knows his business and is building halls exactly the way he wants them to sound. But his concept of sound seems to be wrong, and he has now become accustomed to hearing music in his own halls. He and the players do not seem to be communicating and the reason clearly seems to be that they, are hearing with entirely different concepts of how the instruments should sound7.

 

III. THE FLAW

The basic flaw of Davies Hall is the use of reflecting surfaces in an attempt to achieve equal disbursement of the sound throughout the entire hall. The Anstendig Institute stands by its belief that it is impossible to build a hall with equally perfect distribution of the sound to all seats. One will always be faced with the necessity of optimizing the sound for the central locations. All that can be attempted with the rest of the hall is to minimize falloff and degradation of the sound in other locations, without degrading the centrally-radiated sound.

 

Using reflecting surfaces to accomplish what the shape and basic plan of the hall should have already accomplished cannot work because reflecting surfaces change the character and the equalization of the sound. Unlike light, sound does not retain its original character when reflected, even when it is reflected off appropriate, specially designed surfaces. Because of the idiosyncrasies in the way different sound frequencies radiate, the quality and balance of reflected sound is different from that of direct sound. In everyday life, we are used to hearing direct sound. Techniques for playing musical instruments, and the instruments themselves, have been developed in sound environments consisting mainly of direct sound. This is why attempts to improve the evenness of the disbursement of sound throughout an entire hall by using reflecting surfaces do not work. The sound may measure evenly loud all over the hall, but the character of the sound is degraded and distorted because the overtone structure and other qualities such as the focus of the tone are changed.

The Amadeus Quartet and other visiting artists have stated that Davies Hall was not unpleasant to play in. The conclusion must be that the sound onstage is different from the sound in the auditorium (it is even different in various parts of the auditorium). The conductor, in particular, has the best-sounding location in that he hears the most direct sound in relation to reflected sound and therefore remains unaware of the acoustic distortions of the music. Also, these visitors, who have minimal time to set up and become familiar with the hall, were not laboring under the same burdens of adjusting their playing to the problems as those familiar with Davies Hall. Few musicians whose careers depend on bookings in the concert circuit can allow themselves to be openly critical.

The only way to achieve reasonably equal sound-quality for all members of an audience would be to build a hall larger than desired, and then use only the more central portions.

IV. THE DANGERS

The Anstendig Institute currently sees no point in attending concerts in Davies Hall. Its members do not do so except to evaluate the acoustic, and then they seldom remain for a whole concert.

 

There are dangers, especially for the unsuspecting public, of becoming conditioned to wrong sound qualities. Most of the public has already been conditioned to accept as normal the distorted sound quality of most sound reproduction, especially that of records. For this reason, one should approach all reports of new concert halls with excellent acoustics with the utmost possible skepticism. Most likely, their acoustics reflect what their evaluators are accustomed to hearing in sound systems.

 

Since the expressive quality of recorded performances suffers most from the distortions, the record-listener is not used to hearing as fine an expressive quality as should be the case8. This preconditioning to distorted, unnatural, expressively degraded recorded sound accounts for much of the lack of discrimination that allows audiences to put up with the concerts in Davies Hall. In much the same way that the distortions of recorded sound keep the listener from hearing the true expressive content of the music, the distortions in Davies Hall disturb the content of the performance. This accounts for the fact that so many of the audience do not realize that anything is wrong with the concerts. They simply are not used to hearing music when everything is happening wondrously as it should, a phenomenon that does not happen all the time even under perfect circumstances, and certainly cannot happen under the adverse circumstances of this hall.

 

In evaluating the acoustic of Davies Hall, there is now the confusion that the players, in changing their manner of playing, have made it impossible to be sure just how much of the present sound quality, including the apparent improvements, is due to their playing and how much is due to the hall's acoustic. This is an untenable situation because, even if the sound were improved, the players are playing in an incorrect manner. They should go back to playing their instruments in a classically correct manner, the way they would play them under optimal circumstances. This may be difficult to achieve, but it will quickly become apparent whether or not the acoustic is viable. If it is not, they should not play there until it is corrected. To do otherwise would be artistically indefensible and a disservice to art and the public. The players and the hall are there to create and elevate the audience into the finest possible artistic experiences. At present, the strangeness of the acoustic is an anomaly that prevents real music-making, falsifies the composer's intentions, and conditions the audience to accept bad qualities of sound and interpretation.

Footnotes

1 The thickness of the sound is actually present in all registers but because most instruments play in the musical middle register (corresponding to the middle of the piano keyboard), there is a massing of overtones there. This effect resembles the massing of overtones in sound reproduction as described in our paper "The Massing of Overtones in Sound Reproduction".

2 0ur paper "Concert Hall Acoustics" clarifies the universal misunderstandings about reverberation.

3 We refer the reader to our papers on acoustics and equalization for full explanation of these problems.

4 "The Disaster in Modern Concert Hall Design" explains how the distortions prevalent in sound reproduction have influenced the design of concert halls.

5 This phenomenon is explained in our paper "Sound Equalization in Relation to the Way We Perceive Sound".

6 This is explained in our paper "Concert Hall Acoustics".

7 0ur Institute's other papers on sound equalization and acoustics deal with some of the reasons for misconceptions about what music should sound like, the main reason being the prevalent distortions in unequalized sound reproduction.

8 This is described in our papers on sound equalization and in "Stereo: A Misunderstanding".

 

The Anstendig Institute is a non-profit, tax-exempt, research institute that was founded to investigate stress-producing vibrational influences in our lives and to pursue research in the fields of sight and sound; to provide material designed to help the public become aware of and understand stressful vibrational influences; to instruct the public in how to improve the quality of those influences in their lives; and to provide the research and explanations that are necessary for an understanding of how we see and hear.